Below concept map is a representation of the present author’s understanding of Churchman’s general systems approach, mostly based on the description of the nine categories in ‘The systems approach and its enemies’ (Churchman 1979, 79-100). Reading a concept map of this nature is not something most people are used to. It looks like a plate of spaghetti. Some of the arrows are black and represent the general structure of S, which is any particular activity of humans, in organizations or otherwise, conceived as a system. Other arrows are light grey. They represent aspects of S, the importance of which is easier to understand once the general structure is in place. The first explanation will be done chunk by chunk, later to be followed by ‘inter-chunk’ clarifications. The four chunks in a way describe what soft systems thinking is about: using a general systems approach for planning the management of change to add value. And all this within certain limits: later on I will add a fifth, ‘constraints chunk’ for the remaining three categories left out of below concept map to avoid overwhelming the reader.
A good place to start is in the planning chunk. The systems approach is based on the idea that human activity is best conceived as a system. The concept of ‘system’ is complex, because humans and organizational settings are very diverse and complex themselves. The general systems approach applies to all human activity, both individual and social in all sorts of organizational settings, understood broadly, from family to state, from ideology to business, and from school to parliament. The conceptualization of such a human activity system is done by one or more persons in the role of a designer and is called a plan. It can and must be called a plan, because human activity is characterized by serving a purpose. This is a key link to the value chunk, but we will first deal with management. The inverted comma’s around ‘system’ indicate that there is no such thing as a perfect conceptualization. All meaning, structure or pattern is provisional and subject to debate. So we do wise to ask: What should be the purpose of the system? Who should the client be? And so on and so forth (Churchman 1977, 3).
The management chunk is about the management of the system components, which are subsystems that make up the system and that are designed to cluster activities geared to achieving cluster-specific system objectives. The system is open so it operates in an environment, which is beyond the control of the decision-maker, but may contain environmental factors that cannot be ignored. The boundary of the level of control is a design question as is the level of performance of the system. The components coproduce measures of performance, both enabled and constrained by the environment. These measures or indicators are used by the decision-maker to decide whether things go according to plan. If they do not, the decision-maker may allocate more or fewer resources (including human resources), stop or scale down the plan, or send a designer back to the drawing table. Most of the time, decision-makers prefer the first option, especially in the form of small, incremental steps.
The next chunk is the one dealing with change, the transformation chunk. The intended change is described in terms of objectives, both short-term and long-term. The objectives are established by the designer. They usually reflect some kind of ideal. Ideals can only be pursued in an approximate manner. This is most readily understood by looking at so-called input-output models, which is a common approach to systems of all kinds. In go people and money and out come products or services. Or, when we look at engines: in goes fuel and out comes mechanical energy. The ideal is that the energy value of the fuel is equivalent to the value of the mechanical output energy. This raises the question of efficiency. Churchman gives a description of the input-output model in Chapter 5 of The Systems Approach (1968, 61-78). Another question is that of trade-off between ideals. Because input-output models are relatively simple, they can be very useful in the early stages of system design. They are not very good in handling uncertainties.
This brings us to the fourth or value chunk. Churchman developed his general systems approach to increase the likelihood that human activity will produce some good of some kind. The Wall Street Crash of 1929, the ensuing Great Depression, and the Second World War made him acutely aware of the absence or deficiencies of guarantees in administration, economics and diplomacy that in the end the human condition will improve. Another problem is that of implementation. It is one thing to design interesting plans that address serious issues, but quite another to convince decision-makers that it should be executed. Some of the most frequent reasons for non-implementation are political, strategic, bureaucratic, cultural, or social in nature. To ensure implementation, the designer must in some way activate the decision-maker.
It is important to note that among the nine categories, there are three so-called role categories: the client, the designer, and the decision-maker. In the words of Nelson (2003, 465), one of Churchman’s students, this “focus on people as the scaffolding of a system” differentiates the general systems approach from any other approach. Normally, people are “merely a set in a classification of elements,” but in the systems approach the emphasis is shifted to the functions served by people in a system. In Churchman’s “model one or more individuals can fill a particular role, or the same person can fill different roles at the same time.” This can be represented by a map. An example is the benefit-cost map for the “client” category to trace out where the benefits and costs go. In a business, customers enjoy the benefits of a service or product, but there are other clients, too: workers who receive a salary, manager who receive a higher salary and a bundle of perks, while shareholders receive dividends or higher share prices. The production of an influence map for decision-makers follows the same principles. The idea of role categories is one example of the ways in which the systems approach illustrates the pragmatist tradition “that – very broadly – understands knowing the world as inseparable from agency within it” (Legg and Hookway, 2019).
Five grey spaghetti arrows have not been explained yet. Let us start with the concept map proposition “transformation defines purpose”. This means that the purpose of an activity (or “why”) can be usefully clarified by stating how the purpose is planned to be realized. In other words, by stating the objectives. The next proposition is “client is standard for measures of performance”. Adequate measures of performance or indicators are often very difficult to design in a way that the decision-maker can use them in his management. The indicators must measure the increase in value as experienced by the client. The measures must also integrate the objectives in such a way to enable the decision-maker to stay on track to completion of the transition. This is summarized in the the next proposition, where the “decision-maker adaptively manages components.” The final two propositions centre around the concept of mission: “plan contributes to mission” and “mission guides decision-maker.” The mission of an organization for which a decision-maker is responsible is a “statement of purpose: what the organization seeks to achieve over the long term. [… It] offers a pointer to the overall direction in which strategy will take the organization” (Grant 2008, 21). Mission statements are not always very clear, but they can be derived from the objectives of the plans the decision-maker chooses to implement.
A simple way of operationalizing the general systems approach is described in “Value Distribution Assessment Of Geothermal Development In Lake County, Ca” (Churchman, Nelson and Eacret, 1977). Nine basic questions do the job: (1) Who should the client be? (2) What should be the goals of the system? (3) Should there be a measure of performance for the system? (4) Who should the decision-makers be? (5) What components of the system should the decision-makers control? (6) What should the environment of the system be? (7) Who should be the planners of change in the system? (8) How should plans be implemented? (9) What should be the design of the control of the implemented
plan? Note that these questions are in the “should” or “ought” mode. For the sake of comparison they could also be asked in the “is” mode. Instead of these 9 questions, one could also turn the 27 propositional statements of the concept map into 27 “is” and 27 “ought” questions. Considering that each question can be rephrased in many different ways, the total number of questions for one’s inquiry may easily exceed 100. These are not random questions, they are strongly inter-related and must also be considered in their inter-relatedness. That’s the power of Churchman’s general systems approach. It makes me think of Heinz von Foerster’s ethical imperative. Enjoy but don’t get lost.